mathieu/os-rust
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Match simd_relaxed_fma documentation to fmuladd intrinsic

Browse source
This commit is contained in:
Caleb Zulawski 2024-11-23 19:17:27 -05:00
parent e73e9f9af2
commit 402f6a3530
1 changed files with 7 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

10
library/core/src/intrinsics/simd.rs
View file

@ -612,10 +612,14 @@ extern "rust-intrinsic" {
#[rustc_nounwind] #[rustc_nounwind]
pub fn simd_fma<T>(x: T, y: T, z: T) -> T; pub fn simd_fma<T>(x: T, y: T, z: T) -> T;
/// Computes `(x*y) + z` for each element, with unspecified rounding. /// Computes `(x*y) + z` for each element, non-deterministically executing either
/// a fused multiply-add or two operations with rounding of the intermediate result.
/// ///
/// This may be equivalent to `simd_fma`, or it may relax to rounding each /// The operation is fused if the code generator determines that target instruction
/// operation if that's more efficient. /// set has support for a fused operation, and that the fused operation is more efficient
/// than the equivalent, separate pair of mul and add instructions. It is unspecified
/// whether or not a fused operation is selected, and that may depend on optimization
/// level and context, for example.
/// ///
/// `T` must be a vector of floats. /// `T` must be a vector of floats.
#[cfg(not(bootstrap))] #[cfg(not(bootstrap))]